summaryrefslogtreecommitdiffstats
path: root/talloc.c
Commit message (Collapse)AuthorAgeFilesLines
* talloc: change talloc destructor signaturewm42013-10-131-13/+5
| | | | | | | | | | | | | Change talloc destructor so that they can never signal failure, and don't return a status code. This makes our talloc copy even more incompatible to upstream talloc, but on the other hand this is preparation for getting rid of talloc entirely. (The talloc replacement in the next commit won't allow the talloc_free equivalent to fail, and the destructor return value would be useless. But I don't want to change any mpv code either; the idea is that the talloc replacement commit can be reverted for some time in order to test whether the talloc replacement introduced a regression.)
* talloc: fix strndup group of functionswm42012-10-121-2/+2
| | | | This bug has been fixed years ago in upstream talloc.
* talloc.[ch]: remove "type safety" hack that violates C typesUoti Urpala2011-08-191-1/+2
| | | | | | | | | | | | The destructors used by talloc take a "void *" first parameter. However talloc.h had a #define hack that treated the destructor as a function taking first parameter of type "typeof(ptr)" where ptr is the pointer the destructor is set for. I suppose this was done to add some kind of "type safety" against adding a destructor expecting another type of pointer; however this hack is questionable and violates the real C level typing. Remove the hack from the header and adjust talloc.c to avoid a warning about a C type violation that became visible after removing the hack.
* talloc.c: Update to match current upstream ("likely" macro definitions)Uoti Urpala2008-12-271-2/+10
| | | | | | Use the current macro definitions for likely/unlikely from Samba. The old version lacked parentheses around the non-GCC alternative, but there are no uses where this would actually make a difference.
* Make talloc abort() instead of returning NULLUoti Urpala2008-04-231-9/+9
| | | | | | Replace (hopefully) all cases where normally successful allocations could return NULL with abort(). This should allow skipping most checks on allocation return values.
* Hardcode feature checks in talloc.cUoti Urpala2008-04-231-1/+27
| | | | | | Original talloc build system used autoconf to check for features, most of which were standard C headers. Assume those always exist. Always use a workaround for the one non-standard feature (strnlen).
* Add the talloc memory allocatorUoti Urpala2008-04-231-0/+1724
Copy talloc.c and talloc.h from Samba (last changed 2008-04-17 in commit 7b9a647ebbbe9ec9e1b82b42e3a8916396f91273).